Neural crest stem cells give rise to multiple lineages including the peripheral nervous system and melanocytes. Many birth defects, diseases and cancers including melanoma are caused by defects in neural crest expressed genes. The expression of neural crest genes is controlled by a complex transcriptional regulation. One of the transcription factors needed for appropriate expression is SOX10. The target genes of SOX10 may themselves be essential for proper neural crest development. To identify potential SOX10 target genes, we have analyzed the effects of altered SOX10 expression occurring in melanoma samples on the expression of multiple genes simultaneously using cDNA expression microarrays. We have and are continuing to develop sets of tools to discover and analyze genes that are involved in neural crest-derived melanocyte development and disease. We used cDNA microarray cluster analysis comparing RNA from twenty cell lines from varying stages of neural crest development and melanoma. The cDNA microarray contained over 4500 human melanocyte expressed ESTs; over 2000 mouse ESTs with similarity to known genes; and control genes with important functions in NC-M development and melanoma. Analysis of the clustered expression profiles identified two blocks of cDNAs; each containing neural crest control cDNAs as well as cDNAs that have not been attributed to development or disease. We have identifed that one of the genes is altered in a mouse albinism gene chocolate which also has eye defects. We have identifed several other genes that are early markers of neural crest development. We have also developed an efficient method for functional analysis of these genes by over-expression in immortalized NC-Ms in vitro and melanoblasts in vivo. This involves using a retrovirus system to direct overexpression of genes into defined lineaegs in mice and in cell culture. We have found that several of these genes have expression pattersn of specific neural crest derivatives and are mutated in neural crest diseases. We have also found that overexpression of genes can limit the potential of the neural crest stem cells. We have detailed the process by which WNT signaling specifies neural crest cells to melanocytes. We have generated libraries from melanocytes and used them to annotate the genome of the mouse. This will help in our identification of human disease loci that have melanocyte defects.[unreadable] We have sued viruses to understand heirarchies between transcription facotrs and gene products.[unreadable] Using gene expression and in situ comparasin and comparative sequence analysis we are defining the transcriptional language needed for melanocyte development and undersatnding how this is altered in diseases such as melanoma.